HR: 08:45h
AN: G21D-04    [Abstracts]
TI: Monitoring Southern African Rainfall Season Utilizing Growing Regions
AU: Husak, G J
EM: husak@geog.ucsb.edu
AF: University of California, Santa Barbara Department of Geography, Dept. of Geography University of California, Santa Barbara, CA 93106 United States
AU: * Magadzire, T
EM: tmagadzire@fews.net
AF: University of California, Santa Barbara Department of Geography, Dept. of Geography University of California, Santa Barbara, CA 93106 United States
AB: Variability in timing and amount of rainfall during the growing season in southern Africa can have a dramatic impact on livelihoods in the region. This research integrates satellite model rainfall amounts with expectations for the remainder of the season to provide an envelope of likely outcomes for different growing regions. Satellite information combined with station observations combine to make the African Rainfall Climatology (ARC), which is used to estimate the start of season (SOS) and monitor the season-to-date rainfall accumulations at a pixel level. The Collaborative Historical African Rainfall Model (CHARM) - a 36-year climatology based on available station fields, global climate models and an orographic component - is used to estimate various scenarios for the remainder of the season. The season length is defined by location specific length of growing period provided by the Southern African Development Community (SADC). Once the SOS is observed according to the ARC, seasonal accumulations for each pixel begin and are evaluated at a dekadal interval. These accumulations can be compared to historical accumulations after an equal number of dekads to evaluate the progression of the season as a percentage of historical season-to-date totals for each pixel. Rainfall accumulations for the remainder of the growing period can be tallied for each year of the CHARM dataset, and Gamma probability distribution parameters can be fit to these values. Using these distribution parameters, it is possible to evaluate scenarios for the remainder of the season and combine them with the accumulations from the ARC to arrive at total rainfall accumulated during a growing period. Analysis of these totals can be compared with long-term mean accumulations for the growing period to estimate how crops will fare relative to past performance. Evaluation of various wet and dry scenarios for the remainder of the season, defined here as the 80th percentile and 20th percentile, provide an envelope of expectation for seasonal totals. Early in the season the envelope will be quite wide, but as the season progresses the range provided by these scenarios converges on the final seasonal total. Aggregating these results over different growing regions provided by SADC leads to a regional summary of precipitation conditions for critical crop growing areas in southern Africa. Similar techniques to those presented here could be used as input into various crop and hydrologic models.
DE: 0468 Natural hazards
DE: 0480 Remote sensing
SC: Geodesy [G]
MN: Fall Meeting 2005